Current-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: USB umass hard drive "failed to create xfers" when attaching

On 2020/02/17 21:33, Nick Hudson wrote:
> On 17/02/2020 16:23, Paul Goyette wrote:
> [...]
>>> More info...
>>> First, this is on a amd64 system, witwh 8core/16thread and 128GB of RAM.
>>> On IRC it was suggested (thanks, maya!) that the error message might be
>>> related to memory fragmentation.  I didn't believe it (given how much
>>> RAM I have), but a quick check with top(1) showed that I had more than
>>> 100GB of 'file cache' active.  So, I unmounted all my development trees
>>> (to force the cache to get flushed).  Sure enough, I am now able to
>>> successfully mount the USB drive!
>>> So, sounds like "something somewhere isn't quite right (tm)".  I would
>>> have expected a memory allocation failure to automatically trigger some
>>> mechanism to reclaim some of the file cache...
>> And, based on more discussion, this would seem to be a kernel virtual
>> address fragmentation issue, and not related to physical memory being
>> available.  The concensus on IRC is that this is a bug in the xhci(4)
>> driver.
> Yes and no.
> xhci could do better and support multiple DMA segments, but uvm could
> also help.

To access each DMA segment in usb_dma_t *dma, should it be processed
like this?

for (i = 0; i < dma->udma_block->nsegs; i++) {
	bus_dma_segment_t *ds = &dma->udma_block->map->dm_segs[i];

	process_segment(ds->ds_addr, ds->ds_len);

> Nick

Home | Main Index | Thread Index | Old Index